T’OUNGT’OUNG PAO PAO T’oung Pao 94 (2008) 49-114 www.brill.nl/tpao Another Look at Early Chan: Daoxuan, Bodhidharma, and the Three Levels Movement Eric Greene University of California, Berkeley Abstract As one of the earliest records pertaining to Bodhidharma, Daoxuan’s Xu gaoseng zhuan is a crucial text in the study of so-called Early Chan. Though it is often thought that Daoxuan was attempting to promote the Bodhidharma lineage, recent studies have sug­­­­gested that he was actually attacking Bodhidharma and his later followers. The present article suggests that such readings are incor ­­rect and that Daoxuan was in fact attacking the followers of the Three Levels (Sanjie) movement founded by Xinxing, whose role in defining the meaning of chan during the seventh century has not been sufficiently appreciated. Résumé Étant un des premiers documents à parler de Bodhidharma, le Xu Gaoseng zhuan de Daoxuan constitue une source cruciale pour l’étude du “Chan primitif”. Même si l’on considère le plus souvent que Daoxuan s’efforçait de promouvoir la lignée de Bodhidharma, plusieurs études récentes suggèrent qu’en fait il attaquait Bodhidharma et ses héritiers. L’auteur suggère que cette interprétation n’est pas correcte: en réalité Daoxuan s’attaquait aux adhérents du mouvement des “Trois niveaux” (Sanjie) fondé par Xinxing, dont la contribution à la définition du sens duchan pendant le viie siècle n’a pas encore été suffisamment appréciée. Keywords Buddhism, Chan, Daoxuan, Xinxing, Sanjie movement © Koninklijke Brill NV, Leiden, 2008 DOI: 10.1163/008254308X367022 50 E. Greene / T’oung Pao 94 (2008) 49-114 In the words of Bernard Faure, “Although Bodhidharma’s biography is obscure, his life is relatively well known.”1 Indeed, early records of this monk are so vague, and later hagiography embellishes him so extra­­vagantly, that the best approach seems to be, as Faure ultimately suggests, that we treat Bodhidharma not as an individual but as a textual paradigm. In this view, Bodhidharma’s importance lies largely in his function as a literary trope. But what kind of trope exactly? It would be a mistake, I submit, to take an understanding of Bodhidharma based on narrative material spanning several centuries and apply it uncritically to the very earliest of our sources. Yet, as I will argue in this article, some scholars have misread one key document pertaining to Bodhidharma in precisely this way, albeit with different results and to different ends. Correcting this mistake, moreover, may provide a new perspective on the develop- ment of chan (and Chan) in seventh-century China. To clarify what I mean by this, a few definitions are in order con- cerning the problematic word chan 禪. In its most basic meaning chan is a transliteration of the Sanskrit word dhyāna, which, while referring on the one hand to specific stages of yogic attainment, is also used more generally for what in English we would likely call “meditation”;2 in China the monks and nuns who specialized in this practice came to be known as chan masters (chanshi 禪師). By the Song dynasty, however, the title “chan master” usually referred to a special coterie of enlightened Buddhist masters (the “Chan lineage” chanzong 禪宗) who held themselves to be spiritual heirs of the then famous Bodhidharma. These masters and their characteristic teachings form what is usually referred to in English as “Chan Buddhism,” and it is in this meaning that I will use the term Chan (capital C). The notion of a special lineage descending from Bodhidharma seems to have first appeared only in the late seventh century, more than a 1) Faure 1986: 187. I would like to express my thanks to Robert Sharf, Koichi Shinohara, Chen Jinhua, Timothy Barrett, Jamie Hubbard, Robert Ashmore, and the anonymous reader for T’oung Pao, all of whom read drafts of this paper and offered numerous helpful comments. Chen Jinhua was particularly generous with his time, though in the end I have not followed all of his suggestions. Mistakes that remain as a result are entirely my own. 2) But see Spoonberg 1986 on the difficulty of finding a Buddhist correlate to the word “meditation.” E. Greene / T’oung Pao 94 (2008) 49-114 51 hundred and fifty years after Bodhidharma’s death. Thus, from a historical and analytical point of view, there are many figures, such as Bodhidharma himself and his immediate disciples, whom we should not consider as “Chan masters” even though they were integral to the mythology of the later Chan tradition. From the very beginning, however, Chan as an ideology seems to have arisen from monks who were indeed known in their time as chan masters (chanshi)—that is, meditation masters—or at least, to have looked back to them as its founders. We are thus confronted with a distinction that is lexically obscure yet analytically crucial, for early mediation masters do not necessarily have anything in common with later Chan, and later Chan did not necessarily emphasize meditation. How and by whom the notion of a special transmission from Bodhi- dharma was developed is a complex issue, which mostly lies beyond the scope of this article. But I have dwelled on these definitions because part of what I am interested in is how chan as a concept and chanshi as a title came to mean something new while never fully shedding their old associations. Thus, in what follows I will argue that certain groups (such as the Three Levels movement) helped to change the meaning of chan during the seventh century; yet I am not making the claim that these groups had a direct historical connection with early Chan: rather, my intention is to suggest that early Chan, to the extent that it referred to itself as “Chan” and its proponents as “Chan masters”, was building on a gradual evolution in the meaning of the word chan and in the social role of chan masters, which had begun long before. Bodhidharma One way to study this evolution is to look for concrete situations where the term chan was a site of contestation, where arguments were made as to who should count as a true practitioner of chan, whatever exactly that meant. And this is precisely what happens in one of the oldest sources to mention the monk Bodhidharma, Daoxuan’s 道宣 (596–667) Continuation of the Biographies of Eminent Monks (Xu gaoseng zhuan 續高僧傳). In the concluding summary to the section on “chan prac- titioners” (xichan 習禪), Daoxuan presents his overall views on chan practice and his evaluation of the most eminent chan practitioners of recent times. This summary, which following recent precedent I will 52 E. Greene / T’oung Pao 94 (2008) 49-114 call the Evaluation of the Chan Practitioners (Xichan lun 習禪論), is interesting because it is there—rather than in the biographies proper, where he is often simply reworking other material such as mortuary inscriptions—that Daoxuan directly presents his own views and opi- nions.3 Because the Evaluation of the Chan Practitioners mentions Bodhi- dharma it has naturally drawn the attention of scholars of early Chan. Yanagida Seizan, for example, believed that this text revealed Daoxuan’s appreciation of the unique character of Bodhidharma’s Mahāyāna wall- contemplation (Dacheng biguan 大乘壁觀) meditation methods, which he praised as being of the highest merit. According to Yanagida, this “clearly reflects the influence of the new Chan/chan movement of the Bodhidharma lineage at that time”.4 Yet, while it is no doubt true that later followers of the Bodhidharma lineage would view their founder’s teachings as a new and unique form of practice, it is far from clear that Daoxuan himself held such a view. A full reading of the Evaluation of the Chan Practitioners suggests, rather, that his actual interests were the generally more well-known meditation masters of the day, such as Sengchou 僧稠 (480–560), whom he repeatedly praises as exemplars of proper chan practice.5 3) This is not to say that Daoxuan was not indirectly expressing his own views in the selection and editing of his sources, and we will see examples of this below. On the compi- lation process of monastic biographies, see Shinohara 1988. 4) Yanagida 1967: 14–15. See also Yanagida 1970: 175–176. Because Yanagida is not ma- king an analytical distinction between Chan and chan, I have here used both terms. 5) It has long been known that Sengchou was one of the most important Buddhist monks of the northern Qi 北齊 (550–577); though he is listed as a chan master of great renown, until recently little could be said concerning what kind of practice he may have advocated. A few texts attributed to him have been found on a Dunhuang manuscript, but there has been disagreement over their authenticity. Yanagida (1963, 1970) believed that these texts were all later Northern Chan works. Jan Yün-hua then published a series of articles arguing for their authenticity (Jan 1983a, 1983b, 1990b). More recently Okimoto Katsumi (1997) has supported the same view and has argued that Sengchou actually influenced the teach- ings of Northern Chan. Progress towards a better understanding of Sengchou may be more forthcoming in the light of recent archeological discoveries. Inscriptions from the outside of one of the so-called Xiaonanhai 小南海 caves near modern Anyang directly link this site to Sengchou (his picture is even carved inside the cave), and inscribed sutra excerpts allow us to identify the Nirvana Sutra passages that Sengchou used for meditation, a fact alluded to in his biography in the Xu gaoseng zhuan. On the Xiaonanhai cave see Inamoto 2000 and Yan 1995 and 1998.